Home // ICSEA 2014, The Ninth International Conference on Software Engineering Advances // View article

The Quantification of Risk Factors for Predicting Diabetic Cystoid Macular Edema based on a Hierarchical Approach

Authors:
Eun Byeol Jo
Ju Hwan Lee
Jong Seob Jeong
Byeong Cheol Choi
Sung Min Kim

Keywords: diabetic cystoid macular edema; optical coherence tomography; retina layer; thickness; hierarchical approach

Abstract:
This study suggests a novel risk factor extraction method for retina layers based on a hierarchical approach to distinguish Diabetic Cystoid Macular Edema (DCME) from optical coherence tomography scans. For this, a total of 80 subjects composed of 30 normal and 50 DCME patients were selected. To estimate evaluation variables, a hierarchical approach-based feature extraction algorithm was employed. Evaluation variables were classified into the Total Retina (TR), the Inner Retina (IR), the Photoreceptor Outer Segment (POS), the Outer Retina (OR), the Ganglion Cell (GC), and the Retinal Nerve Fiber Layer (RNFL). The experimental results show the reliable performance of the proposed approach in discriminating DCME from normal subjects. The proposed method could differentiate changes in the thickness of the IR and the POS between the normal and DCME groups. In addition, the most significant degeneration was observed in the central macular area. These results suggest the clinical applicability of the proposed method to the diagnosis of DCME.

Pages: 362 to 367

Copyright: Copyright (c) IARIA, 2014

Publication date: October 12, 2014

Published in: conference

ISSN: 2308-4235

ISBN: 978-1-61208-367-4

Location: Nice, France

Dates: from October 12, 2014 to October 16, 2014